Topic: Claymore's Dual Ethereum AMD+NVIDIA GPU Miner v15.0 (Windows/Linux) - page 950. (Read 6590718 times)

Great. Thanks!!! 

I will release new version with ASM for ETH+SIA and ETH+PASCAL within several hours.

install new drivers

Overall, very pleased with the extra DCR spead in v9.  I've had 2 6x470 rigs running it quite nicely for about a week with very few issues. Went from about 3200 DCR to 4800 on 1 rig set at -dcri 30 and about 43-4400 on the other at dcri 26 due to less PSU headroom on that rig.

Anyway, I just got a 3rd 6x470 rig going on this but it's crashed a few times, but mostly only after I went from 4 to 6 GPUs today. Still have plenty of power headroom but it could be something with those last 2 gpus i just added I suppose... still not sure.

However, what I did notice is that when I look in the log files, it's getting a ton of "DCR weak share - skip" errors. I thought maybe this is what was causing the crashes but when i checked log files on the other 2 i see similar rates of this, so still baffled as to why that 3rd rig is crashing... maybe will switch out risers tomorrow.

But... this got me to thinking... it really seems as though the improved hash-rate with v9 hasn't upped my actual DCR earning by anywhere close to the 50% is speed improvement. I thought maybe this was partially due to increasing difficulty but when I plug the numbers into WhatToMine, im coming up quite a bit shorter than I was with v8, relatively.  With v8, I was with 5-8% of what WhatToMine said, but with v9, that gap is more in the 20-25% lower range. Can I assume this is because while the hash-rate is pumped up, it's also generating many more skipped shares, so the apparent 50% increase in hash-rate is in actuality, much less than that because of the bad shares?

Just curious if anyone else is seeing this issue and if so, what are you doing to try to fix it?

I'm using a batch file. I couldn't see anything in the GUI.

Logs...

09:54:51:949   296c   args: -di 0 -epool stratum+tcp://daggerhashimoto.usa.nicehash.com:3353 -ewal ADDR -epsw x -esm 3 -allpools 1 -estale 0 -dpool stratum+tcp://decred.usa.nicehash.com:3354 -dwal ADDR
09:54:51:949   296c   
09:54:51:949   296c   ÉÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍ»
09:54:51:949   296c   º      Claymore's Dual ETH + DCR/SC/LBC/PASC GPU Miner v9.0      º
09:54:51:949   296c   ÈÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍÍͼ
09:54:51:949   296c   
09:54:52:164   296c   ETH: 1 pool is specified
09:54:52:164   296c   Main Ethereum pool is daggerhashimoto.usa.nicehash.com:3353
09:54:52:164   296c   DCR: 1 pool is specified
09:54:52:164   296c   Main Decred pool is decred.usa.nicehash.com:3354
09:54:52:434   296c   OpenCL platform: AMD Accelerated Parallel Processing
09:54:52:450   296c   OpenCL initializing...

09:54:52:450   296c   AMD Cards available: 2
09:54:52:450   296c   GPU #0: Hawaii, 4096 MB available, 44 compute units
09:54:52:450   296c   GPU #0 recognized as Radeon 290X
09:54:52:450   296c   POOL/SOLO version
09:54:52:450   296c   b315
09:54:52:450   296c   Platform: Windows
09:54:52:481   296c   start building OpenCL program for GPU 0...
09:54:52:982   296c   done
09:54:53:234   296c   GPU #0: algorithm ASM
09:54:53:234   296c   cudaGetDeviceCount failed (35, CUDA driver version is insufficient for CUDA runtime version), probably no CUDA devices
09:54:53:234   296c   No NVIDIA CUDA GPUs detected.
09:54:53:234   296c   Total cards: 1
09:55:02:473   296c   No NVIDIA cards in the list, NVML library will not be used.
09:55:02:536   96c   ETH: Stratum - connecting to 'daggerhashimoto.usa.nicehash.com' <198.11.193.180> port 3353
09:55:02:589   96c   send: {"id": 1, "method": "mining.subscribe", "params": ["MinerName/1.0.0", "EthereumStratum/1.0.0"]}

09:55:02:589   96c   send: {"id": 5, "method": "mining.extranonce.subscribe", "params": []}

09:55:02:604   296c   DUAL MINING MODE ENABLED: ETHEREUM+DECRED

09:55:02:604   296c   ETH: nicehash stratum mode

09:55:02:604   96c   ETH: Stratum - Connected (daggerhashimoto.usa.nicehash.com:3353)
09:55:02:604   296c   "-allpools" option is set, default pools can be used for devfee, check "Readme" file for details.
09:55:02:604   296c   Watchdog enabled
09:55:02:620   296c   Remote management (READ-ONLY MODE) is enabled on port 3333
09:55:02:620   296c   

09:55:02:620   9bc    DCR: Stratum - connecting to 'decred.usa.nicehash.com' <198.11.193.180> port 3354
09:55:02:657   9bc   send: {"id": 1, "method": "mining.subscribe", "params": ["dual 1.0"]}

09:55:02:657   9bc   send: {"id": 5, "method": "mining.extranonce.subscribe", "params": []}


09:55:02:657   9bc    DCR: Stratum - Connected (decred.usa.nicehash.com:3354)
09:55:07:637   96c   checkread timeout
09:55:07:690   9bc   checkread timeout
09:55:12:659   96c   checkread timeout
09:55:12:706   9bc   checkread timeout
09:55:17:677   96c   checkread timeout
09:55:17:724   9bc   checkread timeout
09:55:22:708   96c   checkread timeout
09:55:22:761   9bc   checkread timeout
09:55:24:177   96c   socket was closed remotely
09:55:24:177   96c   ETH: Job timeout, disconnect, retry in 20 sec...
09:55:24:240   9bc   socket was closed remotely
09:55:24:242   9bc    DCR: Job timeout, disconnect, retry in 20 sec...

Any update on other coins getting new Algo?  I am sick of Decred and the lack of good pools.

No DCR pools with 2FA at all.  Other pools that payout directly to wallet are not very reliable.. too many people in them or payouts are delayed. I wish to get back to another coin like Pascal that I can use on Coinotron with the new Algo.

Der Friend

you can help me ?

im have one Rig with sapphire RX 470 4 GB
And another Rig with Sapphire Nitro RX480 4GB

How you reach these speeds???

What values you put in polarios to your modded bios?
and what values you change in sapphre trixx?

obs : im only get 1900 (rx470) and 2150(rx480 nitro)

Thanks
Almir

thanks for all the help guys, well for now its the same problem ill try to continue tomorrow pretty tierd allready. goodnight and thx again;]

Maybe it's the pool, maybe you both are on the same pool..

Great work on 9.0.  I'm seeing a huge increase on DCR on my 6x470 rigs.  I am running a custom bios that I spent way too much time tweaking each card.  All cards are under-volted  as far as possible while stable and all clock speeds are 1200mhz.  The memory speeds vary from 2050 (XFX) to 2250 (Nitro).

This is a mixed XFX / Sapphire Nitro rig. (I am taking a slight hit on ETH -- Typically the Nitro's would do 31.5)

ETH - Total Speed: 183.264 Mh/s
ETH: GPU0 29.844 Mh/s, GPU1 31.292 Mh/s, GPU2 29.830 Mh/s, GPU3 31.280 Mh/s, GPU4 29.729 Mh/s, GPU5 31.288 Mh/s

DCR - Total Speed: 4398.327 Mh/s
DCR: GPU0 716.261 Mh/s, GPU1 751.004 Mh/s, GPU2 715.930 Mh/s, GPU3 750.717 Mh/s, GPU4 713.499 Mh/s, GPU5 750.916 Mh/s

This is a Sapphire "reference/blower" rig

ETH - Total Speed: 183.334 Mh/s
ETH: GPU0 30.412 Mh/s, GPU1 30.741 Mh/s, GPU2 30.406 Mh/s, GPU3 30.639 Mh/s, GPU4 30.746 Mh/s, GPU5 30.391 Mh/s

DCR - Total Speed: 4400.002 Mh/s
DCR: GPU0 729.878 Mh/s, GPU1 737.777 Mh/s, GPU2 729.732 Mh/s, GPU3 735.333 Mh/s, GPU4 737.894 Mh/s, GPU5 729.388 Mh/s

Both of these examples would do 380-420 DCR at a higher drci level on 8.1 -- I did have to bump my power level (afterburner) 5-8% to keep the clocks steady (1200mhz).  Due to heat and power usage I will probably be at 4000 Mh/s when I'm done testing and tuning.


I don't know exactly what was done to achieve such great results -- but if something similar happened to Nvidia cards -- that would be exciting.

Now I'm even getting this message for my own mining...been going smooth for a week. Turned of miner to install new cpu fan...now I get this......anyone?

also if you are not running older versions of amd drivers and you are on windows 10 you might want to download these opencl.dll from here and put into miner folder this worked for me http://cryptomining-blog.com/4519-getting-the-optimum-mining-performance-on-amd-gpus/

and also do what the guy above me said increase virtual memory.

i also dont use any of the setx settings

you need to add more virtual memory..

check here

https://support.lenovo.com/za/en/solutions/ht002951

hope it will help you.

I get this kind of message, i cant believe that its because there is only 1 gpu of 8gb and requirments are 16gb because  i was mining earlier on other pc with 1 card aswell. http://imgur.com/a/08vKG

Hey, just above your last post is mine, copy that into a txt file and after done save as start.bat make sure you select file type all files. then copy and past your BTC address in where it's needed , you do not need to add a worker, this will give mining on NICEHASH the thing is you need to keep going atleast until you have enough fro them to payout, not sure the amount.
 

Rx480 amd

Hey guys, why am i getting devfee cannot connect and retry in 20 seconds red error message? It gets hung up and keeps doing this over and over. I did not disable devfee or anything...I'm mining at ethermine

what are you mining on nvidia or amd

Cheers can someone paste their start.bat info for eth dcr? I'll mine all night for them:D When i start the miner it just crashes i think something is wrong with the info in the file. Thanks guys again.

Hi All, I'm so tiered of reading so I'm jumping to the end to just give my 2cents,

I have learned a lot from a few posts and this is my input if there are still guy's with issues on dropping hashes,

Mining ETH+DCR ( ETH = 60.6 MH/s ) ( DCR = 1757.xx MH/s )
Equipment = 2 x Sapphire RX480 Nitro+ OC Dual UIFE Bios ( YES!!! Modded )
Afterburner V4.3.0
Claymore V9

What I have done is setup the batch file as I mine on Nicehash, yes I might do better on other but I like to get paid in BTC rather than to exchange it later and pay more transfer fees etc etc.

Batch file is as follow.

setx GPU_FORCE_64BIT_PTR 0
setx GPU_MAX_HEAP_SIZE 100
setx GPU_USE_SYNC_OBJECTS 1
setx GPU_MAX_ALLOC_PERCENT 100
setx GPU_SINGLE_ALLOC_PERCENT 100
EthDcrMiner64.exe -dcri 29 -epool daggerhashimoto.eu.nicehash.com:3353 -ewal BTC-ADDRESS.Worker -epsw x -esm 3 -allpools 1 -estale 0 -dbg -1 -dpool decred.eu.nicehash.com:3354 -dwal BTC-ADDRESS.Worker -dpsw x -asm 1
pause

So I have had the same issue with fluctuating hash rate eth 30.5 MH/s to 23 MH/s but as I have read through the posts I have tried using -dcri and by slowly increasing and decreasing as you monitor on the scope of afterburner to find the smoothest line on your wattage usage, "I could get it better but than I loos some of my hashes on ETH",  I got to -dcri 29 and that was the best setting for my setup.. the only thing I just cant seam to understand is that one of my cards use more power than the other 7 to 8 watt more.

well here is a picture of my stats , the drop you see in the pic is a new DAG that got read. 

http://i741.photobucket.com/albums/xx51/ZS3JLR/Mining-rig.jpg

Thanks to Claymore..